FW: Elsevier's ASR

Skip to first unread message

Glenn Hampson

Nov 9, 2023, 1:05:29 PM11/9/23
to osi20...@googlegroups.com

Re-sending this email---the original got stuck in the queue


From: Glenn Hampson
Sent: Thursday, November 9, 2023 9:32 AM
To: osi20...@googlegroups.com
Subject: Elsevier's ASR


Um---who said Elsevier wasn’t going to build an All-Scholarship Repository? This just in (via Richard Poynder’s Twitter feed---as usual):


New York, New York--(Newsfile Corp. - November 8, 2023) - Elsevier, a global leader in scientific information and data analytics, has announced a new offering of enriched and authoritative scientific Datasets to power data applications that solve R&D challenges. Elsevier's Datasets enable researchers, data scientists and practice leaders to answer R&D questions with greater speed and precision across many industries, including life sciences, energy, chemicals and materials, and technology. Use cases span a variety of data science and analytical projects including identifying disease targets using natural language processing, predicting molecule efficacy and toxicity using neural networks, predictive modeling, Key Opinion Leader (KOL) analysis and more. (Source: Elsevier Introduces Authoritative Scientific Datasets to Fuel Innovation and Business-Critical Decisions in Life Sciences, Chemicals and Other Research-Intensive Industries (newsfilecorp.com)


Combine your internal knowledge with our domain-specific curated and enriched data to answer R&D questions with greater precision. Explore data packages and custom options below. (Source: Datasets for Industry | Elsevier)


Using data from open-source scientific literature for your R&D or AI projects can lead to gaps in your data pipeline and hamper these projects during development. In most industries, the majority of relevant scientific literature is not open source or available through free channels. This includes some of the most frequently cited scientific research being published today. (Source: Full-text datasets from Elsevier scientific literature)


I love the idea that someone (finally) is taking the lead on doing something with data. I hate the idea that that someone is Elsevier, or any commercial publisher, because they have most of the big-time data under lock and key. If Elsevier becomes for scientific data extraction what Xerox was for photocopies, then it’s going to take a generation for their market share to drop and for our ambitions for open science to be fully realized. Why? Because making data useful and impactful is what researchers want more than anything else. If commercial publishers can do this effectively, then a major sales pitch for “open” is going to ring hollow. Why put your open data into a big pile on GitHub when you can put it instead where it is actually useful, even if that somewhere is in under lock and key? That rumbling sound you just heard is the globally equity gap getting even wider---first APCs, now this.


The big caveat? There are probably at least two. First, I can’t imagine that Elsevier is the only publisher working on this. So there’s going to be a race to see who can develop the best systems the fastest (although since Elsevier has the deepest pockets and the biggest backlist, their product is going to be the one to watch). And second, I don’t know if even Elsevier has the resources to pull this off. It takes several million dollars per year and a network of dedicated researchers to just keep tabs on the data in one narrow field of study---to track the incoming data, standardize it, curate it, fill in the missing pieces, validate the fixes, build tools to extract value, and so on. Multiply this times 500 different fields and who knows how many subfields. There are going to be overlaps and economies of scale, but it’s going to take an investment of tens to hundreds of millions of dollars annually to make this work. But the profit potential is definitely there---I’ve run the numbers for years. They won’t be giving this access away, but they won’t need to: companies will be lining up to pay for this resource.


It will be interesting to see how this is greeted in the marketplace and how capable the initial prototypes are. Any insights from OSIers into all this would be much appreciated.







Glenn Hampson
Executive Director
Science Communication Institute (SCI)
Program Director
Open Scholarship Initiative (OSI)

Reply all
Reply to author
0 new messages